A Neural Network Identification Technique for a Foil-Air Bearing and its Application to Unbalance Response Analysis

2015 ◽  
Author(s):  
Mohd Firdaus Bin Hassan ◽  
Philip Bonello
Keyword(s):  
Neural Network ◽  
Solution Process ◽  
Training Data ◽  
Response Analysis ◽  
Air Bearing ◽  
State Equations ◽  
Unbalance Response ◽  
Fab Model ◽  
Identification Technique ◽  
Air Film

This paper proposes and studies the non-parametric system identification of a foil-air bearing (FAB) and its application to the frequency-domain nonlinear analysis of a foil-air bearing rotor system. This research is motivated by two advantages: (i) it removes computational limitations by replacing the air film and foil structure state equations by a displacement/force relationship; (ii) if the identification is based on empirical data, it can capture complications that cannot be easily modelled. A numerical model of the FAB is identified using a recurrent neural network (RNN). The training data sets are taken from the simultaneous time domain solution of the air film, foil and rotor equations. The RNN FAB model identified at a single speed is then validated over a range of speeds in two ways: (i) by subjecting it to several sets of input-output data that are different from those used in training; (ii) by using it in the harmonic balance (HB) solution process for the unbalance response of a rotor-bearing system. In either case, the test results using the identified model show good agreement with the exact results obtained using the air film and foil equations, demonstrating the great potential of this method, in the absence of self-excitation effects.

scholarly journals A Neural Network Identification Technique for a Foil-Air Bearing Under Variable Speed Conditions and Its Application to Unbalance Response Analysis

2016 ◽  
Vol 139 (2) ◽  
Author(s):  
Mohd Firdaus Bin Hassan ◽  
Philip Bonello
Keyword(s):  
Neural Network ◽  
Solution Process ◽  
Response Analysis ◽  
Data Sets ◽  
Air Bearing ◽  
Variable Speed ◽  
Unbalance Response ◽  
Fab Model ◽  
Wide Range ◽  
Identification Technique

This paper proposes and studies the nonparametric system identification of a foil-air bearing (FAB). This research is motivated by two advantages: (a) it removes computational limitations by replacing the air film and foil structure equations by a displacement/force relationship and (b) it can capture complications that cannot be easily modeled, if the identification is based on empirical data. A recurrent neural network (RNN) is trained to identify the full numerical model of a FAB over a wide range of speeds. The variable-speed RNN-FAB model is then successfully validated against benchmark results in two ways: (i) by subjecting it to different input data sets and (ii) by using it in the harmonic balance (HB) solution process for the unbalance response of a rotor-bearing system. In either case, the results from the identified variable-speed RNN maintain very good correlation with the benchmark over a wide range of speeds, in contrast to an earlier identified constant-speed RNN, demonstrating the great potential of this method in the absence of self-excitation effects.

scholarly journals Improved Empirical Identification of the Inverse Model of a Squeeze-Film Damper Bearing Based on a Recurrent Neural Network

2018 ◽  
Author(s):  
Ghaith Ghanim Al-Ghazal ◽  
Philip Bonello ◽  
Sergio G. Torres Cedillo
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Structural Model ◽  
Solution Process ◽  
Time History ◽  
Inverse Model ◽  
Training Data ◽  
Squeeze Film ◽  
Squeeze Film Damper ◽  
History Of

Most recently proposed techniques for inverse rotordynamic problems seek to identify the unbalance on a rotor using a known structural model and measurements from externally mounted sensors only. Such non-intrusive techniques are important for balancing rotors that cannot be accessed under operational conditions because of temperature or space restrictions. The presence of nonlinear bearings, like squeeze-film damper (SFD) bearings used in aero-engines, complicates the solution process of the inverse rotordynamic problem. In certain practical aero-engine configurations, the solution process requires a substitute for internal instrumentation to quantify the SFD journal vibration. This can be provided by an inverse model of the SFD bearing which outputs the time history of the relative vibration of the SFD journal relative to its housing, for a given input time history of the SFD force. This paper focuses on the inverse model of the SFD and presents an improved methodology for its identification via a Recurrent Neural Network (RNN) trained using experimental data from a purposely designed rig. The novel application of chirp excitation via two orthogonal shakers considerably improves both the quality of the training data and the efficiency of its generation, relative to an earlier preliminary work. Validation test results show that the RNNs can predict the journal displacement time history with reasonable accuracy. It is therefore expected that such an inverse SFD model would serve as a reliable component in the solution of the wider inverse problem of a rotordynamic system.

scholarly journals Effects of Environmental Factors on Ozone Flux over a Wheat Field Modeled with an Artificial Neural Network

2019 ◽  
Vol 2019 ◽  
pp. 1-11
Author(s):  
Zhilin Zhu
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Environmental Factors ◽  
Crop Yields ◽  
Global Radiation ◽  
Training Data ◽  
Response Analysis ◽  
Ann Model ◽  
Wheat Field ◽  
Artificial Neural

Ozone (O3) flux-based indices are considered better than O3 concentration-based indices in assessing the effects of ground O3 on ecosystem and crop yields. However, O3 flux (Fo) measurements are often lacking due to technical reasons and environmental conditions. This hampers the calculation of flux-based indices. In this paper, an artificial neural network (ANN) method was attempted to simulate the relationships between Fo and environmental factors measured over a wheat field in Yucheng, China. The results show that the ANN-modeled Fo values were in good agreement with the measured Fo values. The R2 of an ANN model with 6 routine independent environmental variables exceeded 0.8 for training datasets, and the RMSE and MAE were 3.074 nmol·m−2·s and 2.276 nmol·m−2·s for test dataset, respectively. CO2 flux and water vapor flux have strong correlations with Fo and could improve the fitness of ANN models. Besides the combinations of included variables and selection of training data, the number of neurons is also a source of uncertainties in an ANN model. The fitness of the modeled Fo was sensitive to the neuron number when it ranged from 1 to 10. The ANN model consists of complex arithmetic expressions between Fo and independent variables, and the response analysis shows that the model can reflect their basic physical relationships and importance. O3 concentration, global radiation, and wind speed are the important factors affecting O3 deposition. ANN methods exhibit significant value for filling the gaps of Fo measured with micrometeorological methods.

Semi-supervised Multi-task Network For Image Aesthetic Assessment

2020 ◽  
Vol 2020 (8) ◽  
pp. 188-1-188-7
Author(s):  
Xiaoyu Xiang ◽  
Yang Cheng ◽  
Jianhang Chen ◽  
Qian Lin ◽  
Jan Allebach
Keyword(s):  
Neural Network ◽  
Semantic Content ◽  
Training Data ◽  
Score Distribution ◽  
Generalization Ability ◽  
Subjective Preference ◽  
Ranking Tasks

Image aesthetic assessment has always been regarded as a challenging task because of the variability of subjective preference. Besides, the assessment of a photo is also related to its style, semantic content, etc. Conventionally, the estimations of aesthetic score and style for an image are treated as separate problems. In this paper, we explore the inter-relatedness between the aesthetics and image style, and design a neural network that can jointly categorize image by styles and give an aesthetic score distribution. To this end, we propose a multi-task network (MTNet) with an aesthetic column serving as a score predictor and a style column serving as a style classifier. The angular-softmax loss is applied in training primary style classifiers to maximize the margin among classes in single-label training data; the semi-supervised method is applied to improve the network’s generalization ability iteratively. We combine the regression loss and classification loss in training aesthetic score. Experiments on the AVA dataset show the superiority of our network in both image attributes classification and aesthetic ranking tasks.

Accurate and Transferable Multitask Prediction of Chemical Properties with an Atoms-in-Molecule Neural Network

2018 ◽  
Author(s):  
Roman Zubatyuk ◽  
Justin S. Smith ◽  
Jerzy Leszczynski ◽  
Olexandr Isayev
Keyword(s):  
Neural Network ◽  
Molecular System ◽  
Computational Cost ◽  
Chemical Properties ◽  
The State ◽  
Molecular Properties ◽  
Training Data ◽  
Dft Methods ◽  
Benchmark Datasets ◽  
Quantum Phenomena

<p>Atomic and molecular properties could be evaluated from the fundamental Schrodinger’s equation and therefore represent different modalities of the same quantum phenomena. Here we present AIMNet, a modular and chemically inspired deep neural network potential. We used AIMNet with multitarget training to learn multiple modalities of the state of the atom in a molecular system. The resulting model shows on several benchmark datasets the state-of-the-art accuracy, comparable to the results of orders of magnitude more expensive DFT methods. It can simultaneously predict several atomic and molecular properties without an increase in computational cost. With AIMNet we show a new dimension of transferability: the ability to learn new targets utilizing multimodal information from previous training. The model can learn implicit solvation energy (like SMD) utilizing only a fraction of original training data, and archive MAD error of 1.1 kcal/mol compared to experimental solvation free energies in MNSol database.</p>

Operational Determination of the Activated Sludge Process Using Neural Networks

1992 ◽  
Vol 26 (9-11) ◽  
pp. 2461-2464 ◽  
Author(s):  
R. D. Tyagi ◽  
Y. G. Du
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Steady State ◽  
Activated Sludge ◽  
Feedforward Neural Network ◽  
Training Data ◽  
Activated Sludge Process

A steady-statemathematical model of an activated sludgeprocess with a secondary settler was developed. With a limited number of training data samples obtained from the simulation at steady state, a feedforward neural network was established which exhibits an excellent capability for the operational prediction and determination.

scholarly journals Automated Ventricular System Segmentation in Paediatric Patients Treated for Hydrocephalus Using Deep Learning Methods

2019 ◽  
Vol 2019 ◽  
pp. 1-9 ◽  
Author(s):  
Michał Klimont ◽  
Mateusz Flieger ◽  
Jacek Rzeszutek ◽  
Joanna Stachera ◽  
Aleksandra Zakrzewska ◽  
...  
Keyword(s):  
Neural Network ◽  
Data Augmentation ◽  
Ct Images ◽  
Policy Transfer ◽  
Training Data ◽  
Intraobserver Variability ◽  
Practical Applications ◽  
Brain Scans ◽  
Rate Policy ◽  
Ct Brain

Hydrocephalus is a common neurological condition that can have traumatic ramifications and can be lethal without treatment. Nowadays, during therapy radiologists have to spend a vast amount of time assessing the volume of cerebrospinal fluid (CSF) by manual segmentation on Computed Tomography (CT) images. Further, some of the segmentations are prone to radiologist bias and high intraobserver variability. To improve this, researchers are exploring methods to automate the process, which would enable faster and more unbiased results. In this study, we propose the application of U-Net convolutional neural network in order to automatically segment CT brain scans for location of CSF. U-Net is a neural network that has proven to be successful for various interdisciplinary segmentation tasks. We optimised training using state of the art methods, including “1cycle” learning rate policy, transfer learning, generalized dice loss function, mixed float precision, self-attention, and data augmentation. Even though the study was performed using a limited amount of data (80 CT images), our experiment has shown near human-level performance. We managed to achieve a 0.917 mean dice score with 0.0352 standard deviation on cross validation across the training data and a 0.9506 mean dice score on a separate test set. To our knowledge, these results are better than any known method for CSF segmentation in hydrocephalic patients, and thus, it is promising for potential practical applications.

Fully automated contrast and non-contrast cardiac view detection in echocardiography a multi-centre, multi-vendor study

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
S Gao ◽  
D Stojanovski ◽  
A Parker ◽  
P Marques ◽  
S Heitner ◽  
...  
Keyword(s):  
Neural Network ◽  
Training Data ◽  
Classification Model ◽  
Validation Dataset ◽  
Funding Source ◽  
Private Company ◽  
Validation Data ◽  
Independent Test ◽  
Model Training ◽  
Confusion Matrices

Abstract Background Correctly identifying views acquired in a 2D echocardiographic examination is paramount to post-processing and quantification steps often performed as part of most clinical workflows. In many exams, particularly in stress echocardiography, microbubble contrast is used which greatly affects the appearance of the cardiac views. Here we present a bespoke, fully automated convolutional neural network (CNN) which identifies apical 2, 3, and 4 chamber, and short axis (SAX) views acquired with and without contrast. The CNN was tested in a completely independent, external dataset with the data acquired in a different country than that used to train the neural network. Methods Training data comprised of 2D echocardiograms was taken from 1014 subjects from a prospective multisite, multi-vendor, UK trial with the number of frames in each view greater than 17,500. Prior to view classification model training, images were processed using standard techniques to ensure homogenous and normalised image inputs to the training pipeline. A bespoke CNN was built using the minimum number of convolutional layers required with batch normalisation, and including dropout for reducing overfitting. Before processing, the data was split into 90% for model training (211,958 frames), and 10% used as a validation dataset (23,946 frames). Image frames from different subjects were separated out entirely amongst the training and validation datasets. Further, a separate trial dataset of 240 studies acquired in the USA was used as an independent test dataset (39,401 frames). Results Figure 1 shows the confusion matrices for both validation data (left) and independent test data (right), with an overall accuracy of 96% and 95% for the validation and test datasets respectively. The accuracy for the non-contrast cardiac views of &gt;99% exceeds that seen in other works. The combined datasets included images acquired across ultrasound manufacturers and models from 12 clinical sites. Conclusion We have developed a CNN capable of automatically accurately identifying all relevant cardiac views used in “real world” echo exams, including views acquired with contrast. Use of the CNN in a routine clinical workflow could improve efficiency of quantification steps performed after image acquisition. This was tested on an independent dataset acquired in a different country to that used to train the model and was found to perform similarly thus indicating the generalisability of the model. Figure 1. Confusion matrices Funding Acknowledgement Type of funding source: Private company. Main funding source(s): Ultromics Ltd.

scholarly journals Robust Approach to Supervised Deep Neural Network Training for Real-Time Object Classification in Cluttered Indoor Environment

2021 ◽  
Vol 11 (15) ◽  
pp. 7148
Author(s):  
Bedada Endale ◽  
Abera Tullu ◽  
Hayoung Shi ◽  
Beom-Soo Kang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Architecture ◽  
Input Data ◽  
Deep Neural Network ◽  
Data Augmentation ◽  
Object Classification ◽  
Training Data ◽  
Gradient Descent Algorithm

Unmanned aerial vehicles (UAVs) are being widely utilized for various missions: in both civilian and military sectors. Many of these missions demand UAVs to acquire artificial intelligence about the environments they are navigating in. This perception can be realized by training a computing machine to classify objects in the environment. One of the well known machine training approaches is supervised deep learning, which enables a machine to classify objects. However, supervised deep learning comes with huge sacrifice in terms of time and computational resources. Collecting big input data, pre-training processes, such as labeling training data, and the need for a high performance computer for training are some of the challenges that supervised deep learning poses. To address these setbacks, this study proposes mission specific input data augmentation techniques and the design of light-weight deep neural network architecture that is capable of real-time object classification. Semi-direct visual odometry (SVO) data of augmented images are used to train the network for object classification. Ten classes of 10,000 different images in each class were used as input data where 80% were for training the network and the remaining 20% were used for network validation. For the optimization of the designed deep neural network, a sequential gradient descent algorithm was implemented. This algorithm has the advantage of handling redundancy in the data more efficiently than other algorithms.

Sign in / Sign up

Export Citation Format

Share Document